As survival rates for at-risk premature infants has increased, retinopathy of prematurity (ROP) has become again a very significant cause of childhood blindness. We have chosen the dog model of oxygen-induced retinopathy (OIR) to study the mechanisms involved in ROP because the neonatal canine retina is only 60% vascularized at birth, comparable to a 28 week gestation human. When this developing system is exposed to hyperoxia for four days, retinal vaso-obliteration occurs that is comparable to human ROP yet the choroidal vasculature is apparently unaffected. When animals are returned to normoxia, florid, persistent retinal neovasculazation and tractional retinal folds occur which are similar to human ROP. We will evaluate new therapeutic agents and delivery systems in this clinically relevant model for ROP. The reagents will be evaluated first in in vitro assays to determine if they are effective in dog and their mode of action: inhibition of adult dog retinal microvascular endothelial cell (ADRME) proliferation, migration, and tube formation. The reagents that are active on dog cells will be incorporated into Elvax, Hydron, and/or biodegradable polymers and the elution kinetics and bioavailability of materials eluted from these polymers will be determined in ADRME proliferation assay. Finally, reagents that show promise for inhibiting angiogenesis in dog will be implanted in vitreous one day alter return to room air from exposure to hyperoxia. In vitro models using angioblasts, and adult and neonatal dog retinal microvascular endothelial cells (cell lines we have in culture) will be undertaken for mechanistic studies of vaso-obliteration and vasoproliferation an for determining conditions that stimulate differentiation of angioblasts into endothelial cells. We will determine which of the vascular cells are most sensitive to hyperoxia and what substance can prevent vaso-obliteration in vitro. In other studies, we will attempt to understand why retinal vasculature constricts so rapidly and undergoes vaso-obliteration while the choriocapillaris remains dilated and resists vaso- obliteration. These studies will provide new insights into OIR and suggest new therapies for this blinding disorder. Some of the therapies proposed target two substances induced by hypoxia, VEGF and adenosine, which we have determined is elevated during vasoproliferation. Other experiments will investigate the effects of antiangiogenic substances currently being considered for clinical trials on tumor angiogenesis (angiostatin, endostatin, squalamine, genistein on neovascularization in this clinically relevant dog model for ROP.